# **AdmirePred**
A method for predicting abundant miRNAs in Exosomes
## Introduction
AdmirePred is a tool for the prediction of miRNA found abundantly in exosomes under normal conditions. It uses similarity-based methods (Basic Local Alignment Search Tool) combined with Extra Tree Classifier built on the best performing composition-based features extracted using One hot encoding and Term Frequency - Inverse Document Frequency. AdmirePred is also available as a web-server at https://webs.iiitd.edu.in/raghava/admirepred. Please read/cite the content about AdmirePred for complete information including algorithm behind the approach.
## Python Package
```
pip install admirepred
```
```
import admirepred
```
It can also be downloaded from - https://pypi.org/project/admirepred/
## Requirements
- scikit-learn=1.6.1
- Pandas
- Numpy
- Joblib
- Argparse
No additional package/tool is required for model = 1 (default model), however for model = 2, please download blast (version - blast: 2.12.0+) from https://blast.ncbi.nlm.nih.gov/doc/blast-help/downloadblastdata.html
## Minimum USAGE
To know about the available option for the standlone, type the following command:
```
admirepred -h
```
To run the example, type the following command:
```
admirepred -f example_seq.fa -o output
```
Here, -f argument is to enter the input file in Fasta format and -o argument is for giving the path to the output directory. By default, the package uses model (-m) = 1 which employs only ML algorithm (Extra Tree Classifier) to classify the miRNA sequences, which generates a prediction file "classification_ML.csv" in the specified output directory. If model (-m) = 2 is selected, then the hybrid model is employed (ML + BLAST) to classify the miRNA sequences, which generates a prediction file "classification_hybrid.csv" in the specified output directory.
## Full Usage
```
usage: admirepred [-h] --file FILE --output OUTPUT [--model MODEL] [--threshold THRESHOLD]
```
```
Please provide following arguments for successful run
required arguments:
--file FILE, -f FILE Path to fasta file
--output OUTPUT, -o OUTPUT Path to output
optional arguments:
--model MODEL, -m MODEL Model selection: 1 for ML only, 2 for ML + BLAST + MERCI (By default model = 1)
--threshold THRESHOLD, -t THRESHOLD Threshold for classification (can be any value between 0-1 for model = 1 (by default = 0.5) and 0-2 for model = 2 (by default = 0.52))
For help:
-h, --help show this help message and exit
```
## Standalone minimum usage
```
python3 admirepred.py -f example_seq.fa -o output
```
## Arguments description
**Input File:** It allow users to provide input in FASTA format.
**Output File:** Program will save the results to this folder
**Model:** User can pick which model to run, model = 1 runs only ML model (ET classifier), whereas model = 2 runs hybrid model (ML + BLAST), by default the tool runs model = 1
**Threshold:** User can provide threshold for classification (can be any value between 0-1 for model = 1 (by default = 0.51) and 0-2 for model = 2 (by default = 0.50))
AdmirePred Package Files
=======================
It contantain following files, brief description of these files given below
INSTALLATION : Installations instructions
LICENSE : License information
README.md : This file provide information about this package
admirepred_et_model.pkl : This file contains the pickled version of model
admirepred.py : Main python program
example_input.fa : Example file contain nucleotide sequences in FASTA format
blast_db : Database for BLAST search
Raw data
{
"_id": null,
"home_page": "https://github.com/raghavagps/admirepred",
"name": "admirepred",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": null,
"keywords": null,
"author": "Akanksha",
"author_email": "akankshaar@iiitd.ac.in",
"download_url": "https://files.pythonhosted.org/packages/b8/9d/102cd83c22b0aad798bc5f9ba1c2a9eec2ce81d996e5b74173880e3045c1/admirepred-1.0.0.tar.gz",
"platform": null,
"description": "# **AdmirePred**\nA method for predicting abundant miRNAs in Exosomes\n## Introduction\nAdmirePred is a tool for the prediction of miRNA found abundantly in exosomes under normal conditions. It uses similarity-based methods (Basic Local Alignment Search Tool) combined with Extra Tree Classifier built on the best performing composition-based features extracted using One hot encoding and Term Frequency - Inverse Document Frequency. AdmirePred is also available as a web-server at https://webs.iiitd.edu.in/raghava/admirepred. Please read/cite the content about AdmirePred for complete information including algorithm behind the approach.\n\n## Python Package\n```\npip install admirepred\n```\n```\nimport admirepred\n```\nIt can also be downloaded from - https://pypi.org/project/admirepred/\n\n\n## Requirements\n- scikit-learn=1.6.1\n- Pandas\n- Numpy\n- Joblib\n- Argparse\n\n\nNo additional package/tool is required for model = 1 (default model), however for model = 2, please download blast (version - blast: 2.12.0+) from https://blast.ncbi.nlm.nih.gov/doc/blast-help/downloadblastdata.html\n\n\n## Minimum USAGE\nTo know about the available option for the standlone, type the following command:\n```\nadmirepred -h\n```\nTo run the example, type the following command:\n```\nadmirepred -f example_seq.fa -o output\n```\nHere, -f argument is to enter the input file in Fasta format and -o argument is for giving the path to the output directory. By default, the package uses model (-m) = 1 which employs only ML algorithm (Extra Tree Classifier) to classify the miRNA sequences, which generates a prediction file \"classification_ML.csv\" in the specified output directory. If model (-m) = 2 is selected, then the hybrid model is employed (ML + BLAST) to classify the miRNA sequences, which generates a prediction file \"classification_hybrid.csv\" in the specified output directory.\n\n## Full Usage\n```\nusage: admirepred [-h] --file FILE --output OUTPUT [--model MODEL] [--threshold THRESHOLD]\n```\n```\nPlease provide following arguments for successful run\nrequired arguments:\n --file FILE, -f FILE Path to fasta file\n --output OUTPUT, -o OUTPUT Path to output\n\noptional arguments:\n\n --model MODEL, -m MODEL Model selection: 1 for ML only, 2 for ML + BLAST + MERCI (By default model = 1)\n --threshold THRESHOLD, -t THRESHOLD Threshold for classification (can be any value between 0-1 for model = 1 (by default = 0.5) and 0-2 for model = 2 (by default = 0.52))\n\nFor help:\n -h, --help show this help message and exit\n\n```\n\n## Standalone minimum usage\n```\npython3 admirepred.py -f example_seq.fa -o output\n```\n\n## Arguments description\n\n**Input File:** It allow users to provide input in FASTA format.\n\n**Output File:** Program will save the results to this folder\n\n**Model:** User can pick which model to run, model = 1 runs only ML model (ET classifier), whereas model = 2 runs hybrid model (ML + BLAST), by default the tool runs model = 1\n\n**Threshold:** User can provide threshold for classification (can be any value between 0-1 for model = 1 (by default = 0.51) and 0-2 for model = 2 (by default = 0.50))\n\n\nAdmirePred Package Files\n=======================\nIt contantain following files, brief description of these files given below\n\nINSTALLATION : Installations instructions\n\nLICENSE : License information\n\nREADME.md : This file provide information about this package\n\nadmirepred_et_model.pkl : This file contains the pickled version of model\n\nadmirepred.py : Main python program\n\nexample_input.fa : Example file contain nucleotide sequences in FASTA format\n\nblast_db : Database for BLAST search\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "A tool to predict abundant miRNA in exosomes",
"version": "1.0.0",
"project_urls": {
"Homepage": "https://github.com/raghavagps/admirepred"
},
"split_keywords": [],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "01b5bef0e08ec7e1f3bcbd3ed4f7fa0305036e0ea20b5ceb2c67a1f72c4973c1",
"md5": "4e0785acaebae8922ec9fc51d5d2f031",
"sha256": "d391082f0223e6b8570f015bd549bc18a4ba500ed6226e654646d9e18638fe88"
},
"downloads": -1,
"filename": "admirepred-1.0.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "4e0785acaebae8922ec9fc51d5d2f031",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.9",
"size": 652168,
"upload_time": "2025-02-03T12:58:47",
"upload_time_iso_8601": "2025-02-03T12:58:47.900090Z",
"url": "https://files.pythonhosted.org/packages/01/b5/bef0e08ec7e1f3bcbd3ed4f7fa0305036e0ea20b5ceb2c67a1f72c4973c1/admirepred-1.0.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "b89d102cd83c22b0aad798bc5f9ba1c2a9eec2ce81d996e5b74173880e3045c1",
"md5": "3406499dd57e86c0544920f6c570d5f1",
"sha256": "719fe9a41ff514d86167a9366d5d8952db6a228afcf73feba3b4083a8bf1a347"
},
"downloads": -1,
"filename": "admirepred-1.0.0.tar.gz",
"has_sig": false,
"md5_digest": "3406499dd57e86c0544920f6c570d5f1",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.9",
"size": 624371,
"upload_time": "2025-02-03T12:58:49",
"upload_time_iso_8601": "2025-02-03T12:58:49.821530Z",
"url": "https://files.pythonhosted.org/packages/b8/9d/102cd83c22b0aad798bc5f9ba1c2a9eec2ce81d996e5b74173880e3045c1/admirepred-1.0.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-02-03 12:58:49",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "raghavagps",
"github_project": "admirepred",
"github_not_found": true,
"lcname": "admirepred"
}
JSON
